Possible Interpretations of Statistics of Simulated Interannual Sea Ice-Atmosphere Interations in Greenland Sea
一个冰气相互作用数值模拟结果的可能统计解释

Statistical relationships between principal component time series of the leading empirical orthogonal functions of several physical quantities(sea surface temperature,sea ice concentration,surface sensible heat flux and so on in winter) in Greenland Sea are analyzed from the last 30 years' result of a 50 years' integration experiment with LASG/NCC (State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics,Chinese Academy of Sciences/China National Climate Center) global coupled model.It is shown that positive(negative) anomalies of sea ice concentration in the former year are connected with negative(positive) anomalies of sea surface temperature in the following year and they have the biggest correlation in the same year.Air temperature at the lowest model layer have the biggest correlation with sea surface heat flux in the same year and positive(negative) anomalies of air temperature are connected with negative(positive) surface sensible and latent heat flux anomalies sent to the air.Both air temperature and air humidity at the lowest model layer have the biggest correlation with sea level pressure in the area covering Icelandic low and Greenland Sea and negative(positive)anomalies of Icelandic low are connected with positive(negative) anomalies of air temperature and humidity.So,in the course of interannual sea-sea ice-atmosphere interaction,the dominant relation is that,adjustments of atmospheric general circulation cause the changes of cloud,air temperature and humidity,which influence the flux exchange through interface of sea and air and give contribution to the variation of sea surface temperature.Variation of sea surface temperature causes variation of ice extent and ice concentration,whereas variation of ice masses can also play discernable influences on sea surface temperature through absorption or release of latent heat.The influence of ice extent variation on sea surface net solar radiation is obvious whereas its further effect on surface air temperature(i.e.,positive ice-albedo feedback effect) is not discernable.It should be pointed out that the sea ice model used here doesn't include dynamic processes.Hence the interaction of sea-sea ice and ice-air deployed in the model are nearly pure thermodynamic in high northern latitudes.If the effects of ice motion on sea-sea ice-air interaction are considered,new features might appear.